Two-hour glucose is a better risk predictor for incident coronary heart disease and cardiovascular mortality than fasting glucose.

Abstract

To assess the predictive value of fasting and 2-h glucose after a 75 g glucose load, with regard to incidence of coronary heart disease and cardiovascular mortality. 6766 subjects from five Finnish cohorts aged 30-89 years were followed up for 7-10 years. Hazards ratios associated with increasing glucose concentrations were homogeneous over studies. Multivariate Cox regression analyses showed that the hazards ratio for one standard deviation increase in 2-h glucose after logarithmic transformation was 1.17 (95% CI 1.05-1.30) for coronary heart disease incidence and 1.22 (1.09-1.37) for cardiovascular mortality. For fasting glucose, they were 1.05 (0.94-1.17) and 1.13 (1.01-1.25), respectively. Inclusion of 2-h glucose in the model based on fasting glucose significantly improved the prediction (P<0.005 for coronary heart disease incidence and P<0.025 for cardiovascular mortality), whereas fasting glucose did not add significant information to the model initially based on 2-h glucose (P>0.10 for both events). In subjects without a prior history of diabetes the association of 2-h glucose with coronary heart disease incidence and cardiovascular morality is graded and independent. The results of our study indicate that 2-h glucose is superior to fasting glucose in assessing the risk of future cardiovascular disease events.